The cluster-indexing method for case-based reasoning using self-organizing maps and learning vector quantization for bond rating cases

2001 ◽  
Vol 21 (3) ◽  
pp. 147-156 ◽  
Author(s):  
Kyung-Sup Kim ◽  
Ingoo Han
2020 ◽  
Vol 6 (1) ◽  
pp. 53
Author(s):  
Fhatiah Adiba ◽  
Nurul Mukhlisah Abdal ◽  
Andi Akram Nur Risal

This study aims to compare the results of the accuracy and speed of the system in diagnosing skin diseases using the case based reasoning (CBR) method with the indexing method and without using indexing. Self-organizing maps (SOM) are used as an indexing method and the process of finding similarity values uses the nearest neighbor method. Testing is done with two scenarios. The first scenario uses CBR without indexing self-organizing maps, the second scenario uses CBR with indexing self-organizing maps. The accuracy of the diagnosis of skin diseases at a threshold ≥80 for CBR without indexing self-organizing maps is 93.46% with an average retrieve time of 0.469 seconds while CBR testing using SOM indexing is 92.52% with an average retrieve time of 0.155 seconds. The results of comparison of CBR methods without using show higher results than using SOM indexing, but the process of retrieving CBR using SOM is faster than not using indexing


2009 ◽  
Vol 57 (7) ◽  
pp. 2763-2769 ◽  
Author(s):  
José S. Torrecilla ◽  
Ester Rojo ◽  
Mercedes Oliet ◽  
Juan C. Domínguez ◽  
Francisco Rodríguez

Sign in / Sign up

Export Citation Format

Share Document